An improved segmentation approach for planar surfaces from unstructured 3D point clouds

The extraction of object features from massive unstructured point clouds with different local densities, especially in the presence of random noisy points, is not a trivial task even if that feature is a planar surface. Segmentation is the most important step in the feature extraction process. In practice, most segmentation approaches use geometrical information to segment the 3D point cloud. The features generally include the position of each point (X, Y and Z), locally estimated surface normals and residuals of best fitting surfaces; however, these features could be affected by noisy points and in consequence directly affect the segmentation results. Therefore, massive unstructured and noisy point clouds also lead to bad segmentation (over-segmentation, under-segmentation or no segmentation). While the RANSAC (random sample consensus) algorithm is effective in the presence of noise and outliers, it has two significant disadvantages, namely, its efficiency and the fact that the plane detected by RANSAC may not necessarily belong to the same object surface; that is, spurious surfaces may appear, especially in the case of parallel-gradual planar surfaces such as stairs. The innovative idea proposed in this paper is a modification for the RANSAC algorithm called Seq-NV-RANSAC. This algorithm checks the normal vector (NV) between the existing point clouds and the hypothesised RANSAC plane, which is created by three random points, under an intuitive threshold value. After extracting the first plane, this process is repeated sequentially (Seq) and automatically, until no planar surfaces can be extracted from the remaining points under the existing threshold value. This prevents the extraction of spurious surfaces, brings an improvement in quality to the computed attributes and increases the degree of automation of surface extraction. Thus the best fit is achieved for the real existing surfaces.     

Study of the optimum time span for distinguishing <Emphasis Type="Italic">Avena ludoviciana</Emphasis> from wheat crop based on their spectral reflectance characteristics

The study to establish the optimum time span for distinguishing Avena ludoviciana from wheat crop based on their spectral signatures was carried out at Student’s Research Farm, Department of Agronomy during 2006–07 and 2007–08. The experimental sites during both the seasons were sandy loam in texture, with normal soil reaction and electrical conductivity, low in organic carbon and available nitrogen and medium in available phosphorus and potassium. The experiment was laid out in randomized block design with four replications and consisting of twelve treatments comprising 0, 10, 15, 25, 50, 75, 100, 125, 150, 200, 250 plants m⁻² and a pure Avena ludoviciana plot (T_max). The results revealed that in all the treatments irrespective of wheat and weeds, the red reflectance (%) value decreased from 34 to 95 DAS (days after sowing) in 2006–07 and 45 DAS to 100 DAS during 2007–08, and thereafter a sharp increase was observed in all the treatments. This trend might be due to increased chlorophyll index after 34 DAS as red reflectance was reduced by chlorophyll absorption. Among all the treatments, T_max (Pure Avena ludoviciana plot) had the highest red reflectance and T₀ (Pure wheat plot) had a lowest value of red reflectance during both the years. The highest value of IR reflectance was obtained at 95 DAS (2006–07) and 70 DAS (2007–08) in all the treatments. IR reflectance of wheat crop ranged between 24.61 and 61.21 per cent during 2006–07 and 27.33 and 67.3 per cent during 2007–08. However, IR reflectance values declined after 95 DAS and 70 DAS up to harvesting during 2006–07 and 2007–08. This lower reflectance may have been due to the onset of senescence. The highest RR and NDVI values were recorded under pure wheat treatment and minimum under pure weed plots. This may be due to dark green colour and better vigor of the wheat as compared to Avena ludoviciana. It was observed that by using RR and NDVI, pure wheat can be distinguished from pure populations of Avena ludoviciana after 34 DAS and different levels of weed populations can be discriminated amongst themselves from 68 DAS up to 107 DAS during both the years of investigation.     

Variability analysis of groundwater levels — AGIS-based case study

The present study has analyzed the variability in depth to water level below ground level (bgl) vis-à-vis groundwater development and rainfall from 1987 to 2007 in agriculture dominated Kaithal district of Haryana state in India. Spatial distribution of groundwater depth was mapped and classified into different zones using ILWIS 3.6 GIS tools. Change detection maps were prepared for 1987–1997 and 1997–2007. Groundwater depletion rates during successive decades were compared and critical areas with substantial fall in groundwater levels were identified. Further, block wise trends of change in groundwater levels were also analyzed. The water table in fresh belt areas of the district (Gulha, Pundri and Kaithal blocks) was observed to decline by a magnitude ranging from 10 m to 23 m. In Kalayat and Rajaund blocks, the levels were found fluctuating in a relatively narrow range of 4–9 m. During 1997–2007, the depletion has been faster compared to the preceding decade. Excessive groundwater depletion in major part of the district may be attributed to indiscriminate abstraction for irrigation and decrease in rainfall experienced since 1998. Changes in cropping pattern and irrigation methods are needed in the study area for sustainable management of the resource.     

Approaches for delineating landslide hazard areas using different training sites in an advanced artificial neural network model

The current paper presents landslide hazard analysis around the Cameron area, Malaysia, using advanced artificial neural networks with the help of Geographic Information System (GIS) and remote sensing techniques. Landslide locations were determined in the study area by interpretation of aerial photographs and from field investigations. Topographical and geological data as well as satellite images were collected, processed, and constructed into a spatial database using GIS and image processing. Ten factors were selected for landslide hazard including: 1) factors related to topography as slope, aspect, and curvature; 2) factors related to geology as lithology and distance from lineament; 3) factors related to drainage as distance from drainage; and 4) factors extracted from TM satellite images as land cover and the vegetation index value. An advanced artificial neural network model has been used to analyze these factors in order to establish the landslide hazard map. The back-propagation training method has been used for the selection of the five different random training sites in order to calculate the factor’s weight and then the landslide hazard indices were computed for each of the five hazard maps. Finally, the landslide hazard maps (five cases) were prepared using GIS tools. Results of the landslides hazard maps have been verified using landslide test locations that were not used during the training phase of the neural network. Our findings of verification results show an accuracy of 69%, 75%, 70%, 83% and 86% for training sites 1, 2, 3, 4 and 5 respectively. GIS data was used to efficiently analyze the large volume of data, and the artificial neural network proved to be an effective tool for landslide hazard analysis. The verification results showed sufficient agreement between the presumptive hazard map and the existing data on landslide areas.     

Entropy-based fuzzy classification parameter optimization using uncertainty variation across spatial resolution

In the past researchers have suggested hard classification approaches for pure pixel remote sensing data and to handle mixed pixels soft classification approaches have been studied for land cover mapping. In this research work, while selecting fuzzy c-means (FCM) as a base soft classifier entropy parameter has been added. For this research work Resourcesat-1 (IRS-P6) datasets from AWIFS, LISSIII and LISS-IV sensors of same date have been used. AWIFS and LISS-III datasets have been used for classification and LISS-III and LISS-IV data were used for reference data generation, respectively. Soft classified outputs from entropy based FCM classifiers for AWIFS and LISS-III datasets have been evaluated using sub-pixel confusion uncertainty matrix (SCM). It has been observed that output from FCM classifier has higher classification accuracy with higher uncertainty but entropy-based classifier with optimum value of regularizing parameter generates classified output with minimum uncertainty.     

Accuracy improvement of ASTER stereo satellite generated DEM using texture filter

The grid DEM(digital elevation model) generation can be from any of a number of sources:for instance,analogue to digital conversion of contour maps followed by application of the TIN model,or direct elevation point modelling via digital photogrammetry applied to airborne images or satellite images.Currently,apart from the deployment of point-clouds from LiDAR data acquisition,the generally favoured approach refers to applications of digital photogrammetry.One of the most important steps in such deployment i...     

Land suitability studies for major crops in Pavagada taluk,Karnataka using remote sensing and GIS techniques

Pavagada taluk of Tumkur district in Karnataka is one of the most backward taluks receiving less than 500 mm annual rainfall. The maximum area of the taluk is under monocropping, reasons for the same were not documented well. The present study was carried out using remote sensing data along with field survey and laboratory analysis for assessing the potentials and limitations of soil. Using the basic information on soil, climate and topography based on the matching exercise between the growth and production requirements of the crop, suitability of soils for groundnut, paddy and finger millet was assessed as per FAO land evaluation. The soil suitability maps were prepared using Arc GIS software. About 48 per cent of the total area was moderate to marginally suitable and 13 per cent of the area was not suitable for both groundnut and finger millet. Lowland areas covering 12 per cent of the area was highly suitable, 15 per cent was moderate to marginally suitable and 20 per cent was not suitable for paddy cultivation.     

Monitoring growth of built-up areas in indo-gangetic plain using multi-sensor remote sensing data

This paper provides an approach for rapid and accurate estimation of built-up areas on a per pixel-basis using a integration of two coarse spatial resolution remote sensing data namely DMSP-OLS and MODIS NDVI. The DMSP-OLS data due to its free availability, high temporal resolution and wide swath was used for regional level mapping of built-up areas. However, due to its low radiometric resolution, the built-up areas cannot be estimated accurately from the DMSP-OLS data. In present study, the DMSP-OLS data was combined with MODIS NDVI data to develop an Human Settlement Index (HSI) image, which estimated the fraction of built-up area on a per pixel basis. The resultant HSI image conveys more information than both the individual datasets. These temporal HSI images were then used for monitoring urban growth in Indo-Gangetic plains during the 2001–2007 time period. Thus, the present research can be very useful for regional level monitoring of built-up areas from coarse resolution data within limited time and minimal cost.     

Assessment of coastal bio-resources of the Palk Bay,India, using IRS-LISS-III data

Coastal resources viz., coral reefs, seagrasses, mangroves, and coastal land features viz., sandy beach, mudflats and salt pan/aquaculture ponds were classified and assessed in the Palk Bay region of the south-east coast of India using IRS LISS III satellite image (1996, 2000, 2002 and 2004). The study recorded an areal coverage of 286.95 ha of reef area during 2004, which is 177.54 ha lesser than that of the reef area of 1996. The reef vegetation composed mainly of seaweeds has gained over 29.44 ha during the same period. Likewise, sand over reef area has also increased alarmingly i.e. 120.34 ha between 1996 and 2004. The seagrass beds of Munaikkadu region of the Palk Bay are comparatively protected and have gained over 7.5 ha between 1996 and 2004. It has been found that both the dense (2.99 ha) and sparse (36.45 ha) mangroves have gained their areal coverage considerably between 1996 and 2004. Whereas in Devipattinam region, many anthropogenic pressures are exerted only on the seagrass resources which has led to the reduction of over 785.5 ha of dense seagrass beds between 1996 and 2004. The study clearly indicated that the resources are under the pressures of low to high threats, especially the coral reefs and seagrasses, if the pressures continue, coastal resources of the Palk Bay may face serious threats of destruction in this part of the Bay in the years to come.     

A bias-free geodetic boundary value problem approach to height datum unification

A geodetic boundary value problem (GBVP) approach has been formulated which can be used for solving the problem of height datum unification. The developed technique is applied to a test area in Southwest Finland with approximate size of 1.5° × 3° and the bias of the corresponding local height datum (local geoid) with respect to the geoid is computed. For this purpose the bias-free potential difference and gravity difference observations of the test area are used and the offset (bias) of the height datum, i.e., Finnish Height Datum 2000 (N2000) fixed to Normaal Amsterdams Peil (NAP) as origin point, with respect to the geoid is computed. The results of this computation show that potential of the origin point of N2000, i.e., NAP, is (62636857.68 ± 0.5) (m²/s²) and as such is (0.191 ± 0.003) (m) under the geoid defined by W ₀ = 62636855.8 (m²/s²). As the validity test of our methodology, the test area is divided into two parts and the corresponding potential difference and gravity difference observations are introduced into our GBVP separately and the bias of height datums of the two parts are computed with respect to the geoid. Obtaining approximately the same bias values for the height datums of the two parts being part of one height datum with one origin point proves the validity of our approach. Besides, the latter test shows the capability of our methodology for patch-wise application.